[Eeglablist] Time-frequency method

[Makoto Miyakoshi]

Dear Ayda,

If you are only interested in the beta band, an easier solution I would
recommend is to apply band-pass filter to leave the beta band, and compute
instantaneous amplitude by applying Hilbert transform in this way:
abs(hilbert(EEG.data)). This should have the best temporal resolution.

If you still want to use newtimef, I suggest you use cycles more than 3,
since EEGLAB 3-cycle Morlet wavelet frequency profile I saw was quite poor.
If you did not like cycle == 6, try 5, 4, ... with due cost.

Makoto


On Mon, Jun 2, 2014 at 12:01 PM, Ayda Ghahremani <ghahremani.aida at gmail.com>
wrote:

> Dear EEGLAB,
>
> I am interested to look at the beta modulation at around 20 Hz in the TF
> plot. Firstly, in the time-frequency analysis, I chose the cycles linearly
> increase from 3 at 3 Hz to 9 cycles at 40 Hz. Then, at 20 Hz it will be
> around 6 cycles if I am not wrong making the resolution around (6×50ms=300
> ms). For my purpose,this time resolution for beta was too low,so I decided
> to choose 3 cycles fixed (as a result wavelet analysis). The frequency
> resolution as a result came down so that you cannot distinguish the higher
> freq anymore. I am not sure about pros and cons of the wavelet in this way.
> I appreciate your help if you guys have some experimental experience with
> it.
>
> Ayda
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20140616/af7de8a6/attachment.html>